"Thorsten Ottosen" <nesotto@cs.auc.dk> wrote in message news:cvkf1m$unc$1@sea.gmane.org...
I'm not sure I like the expression "trees as algorithms". I would like to see different types of iterators which can then be used to implement algorithms.
So I only see trees as containers.
Right-o. At this point, I'm going to take the feedback we've gathered so far (thank you everyone =) and begin the redesign of the core::tree to the most simplistic design I can and try to incorporate as many ideas suggested here that make sense for that basic design. I'll also try to improve the code quality while removing any unneeded basic functionality that exists in the tree currently. The suggestions here have been amazing, please keep them coming. Perhaps as we move forward, we can begin to incorporate more and more ideas into the general tree framework. Additionally, I'm going to try to keep the number of template parameters down to the lowest number possible while still allowing all the functionality we need (template <typename T, typename tree_traits, typename allocator>). As far as the generic base tree, I strongly feel that the more simple it is, the better. As we all seem to be pointing out, we really need to nail down this generic base tree structure and its core functionality. Once that's done (and we agree on how it's done) we can move into the extremely challenging realm of how to use that structure to achieve all possible trees. Thanks to all of you for the great feedback, assistance and support, =) Justin